CFB Betting Edge

2024 · Week 3 · Sun Belt · backtest — what the model predicted vs what happened, and vs the market
Week 3 backtest · Sun Belt · 8 games — what the model predicted vs what happened
Straight up
5–363%
ATS vs close
4–450%
Model margin MAE
8.3
Market margin MAE
8.1
→ market was closer (model can't beat the close ATS)
Results — model read vs the final score + how the closing-line lean held up
Model
pred Michigan 16.5 · actual Michigan 10
winner Michigan
ATS vs close
leaned Arkansas State +22.0Cover ✓
margin err: model 6.5 · mkt 12.0 · closer
Troy @ IowaFinal 2138
Model
pred Iowa 13.0 · actual Iowa 17
winner Iowa
ATS vs close
leaned Troy +23.5Cover ✓
margin err: model 4.0 · mkt 6.5 · closer
Model
pred East Carolina 8.9 · actual App State 2
winner East Carolina
ATS vs close
leaned East Carolina +1.5Miss ✗
margin err: model 10.9 · mkt 0.5 · mkt closer
Model
pred Texas State 4.2 · actual Arizona State 3
winner Texas State
ATS vs close
leaned Texas State -2.5Miss ✗
margin err: model 7.2 · mkt 5.5 · mkt closer
Model
pred Vanderbilt 5.5 · actual Georgia State 4
winner Vanderbilt
ATS vs close
leaned Georgia State +9.0Cover ✓
margin err: model 9.5 · mkt 13.0 · closer
Model
pred South Florida 10.8 · actual South Florida 25
winner South Florida
ATS vs close
leaned Southern Miss +11.5Miss ✗
margin err: model 14.2 · mkt 13.5 · mkt closer
Model
pred Virginia Tech 14.0 · actual Virginia Tech 20
winner Virginia Tech
ATS vs close
leaned Old Dominion +15.5Miss ✗
margin err: model 6.0 · mkt 4.5 · mkt closer
Model
pred Coastal Carolina 16.0 · actual Coastal Carolina 8
winner Coastal Carolina
ATS vs close
leaned Temple +17.5Cover ✓
margin err: model 8.0 · mkt 9.5 · closer
The model beats SP+ but not the closing line (ATS ≈ break-even). “Model−Mkt” is the gap between our projected home margin and the market's implied margin — a read on disagreement, not a profitable edge. Upcoming games have no closing line yet, so the market number marked ~ is the consensus midpoint of the best posted spreads across books.